Abstract
Multi-Processor Systems-on-Chip (MPSoCs) offer superior performance while maintaining flexibility and reusa- bility thanks to software oriented personalization. While most MPSoCs are today heterogeneous for better meeting the targeted application requirements, homogeneous MPSoCs may become in a near future a viable alternative bringing other benefits such as run-time load balancing, task migration, etc. The work presented in this paper relies on a homogeneous NoC-based MPSoC framework we developed for exploring scalable and adaptive on-line continuous mapping techniques. Each processor of this system is compact and runs a tiny preemptive operating system that monitors various metrics and is entitled to take remapping decisions through code migration techniques. This approach that endows the architecture with decisional capabilities permits refining application implementation at run-time accor- ding to various criteria. Experiments based on simple policies are presented on various applications that demonstrate the benefits of such an approach.